• Intel® Graphics Performance Analyzers 2019 R4
  • 12/20/2019
  • Public Content
Contents

Minimizing Overdraw

With the Graphics Frame Analyzer, you can discover which API calls affect specific pixels and minimize overdraw by removing redundant or unimportant calls. For example, lighting effects are typically produced by multiple rendering passes. You can determine whether you can reduce the number of iterations to produce a similar visual effect with higher performance. To detect the sequence of API calls that affects the final color of the pixel:
  1. Select the color buffer of the render target from the Resource List. Graphics Frame Analyzer opens the render target image in the
    Profiling View
    .
  2. Analyze the render target image using different visualization modes available from the drop-down menu:
    • Current
      (default) - shows the render target with modifications. If no experiments are applied, the current state is equal to the original state. In this mode, you can visualize the color distribution through a histogram by clicking the button.
    • Original
      - shows the render target without modifications.
    • Diff
      - shows the difference between the current and original modes.
    • Overdraw
      - visualizes the hit count for each pixel of the render target image. The brighter the color, the more times the pixel was updated. Hover over the pixel to see the hit count below.
  3. In the
    Overdraw
    mode, hover over the pixel you would like to analyze. Graphics Frame Analyzer displays the hit count for the selected pixel below the image.
  4. Click the pixel you would like to analyze. Graphics Frame Analyzer marks the selected pixel with a crosshair and filters the API calls to display only those clear, draw, blit, or texture image update calls that affected this pixel, with the resulting pixel color shown next to each call.
  5. Review the pixel history in the . Graphics Frame Analyzer displays the pixel coordinates, pixel components, the number of times each draw call touched the pixel, and rejected draw calls. You can switch back to the color buffer to better understand the place of the selected pixel in the overall scene.
    TIP
    To find out the reason for draw call rejection, you can tweak the pipeline state for this draw call. For example, disable depth or stencil tests. For draw calls rejected with scissor test or color mask, the
    API Log
    shows the rejection reason automatically.
  6. Analyze the calls that affected the pixel to determine how to get the same or similar results at less cost:
    • If you see that two API calls affecting the pixel are of the same type, you may improve performance by removing one of them. To see how much performance you can gain by removing an API call, select it, and check the
      Disable Event(s)
      check box in the
      Experiments
      pane
      .
    • Detect API calls used by mistake. Remove such calls by checking the
      Disable Event(s)
      check box in the
      Experiments
      pane to see the potential performance improvement.
      NOTE
      You can only perform the
      Disable Event(s)
      experiment if metrics data is available for your target system. Otherwise, the
      Experiments
      pane is hidden.
Scenes with many overlapping objects, or objects with semitransparent textures (such as smoke, fog, or water) often cause a high load on the graphics card and therefore lower frame rates. If you see that the overdraw is an issue, you could try culling primitives prior to rendering, altering the draw order (so that Z-buffer tests will reject many primitives prior to rendering them with an expensive pixel shader), or using other level-of-detail optimizations that simplify the complexity of the scene.

Product and Performance Information

1

Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice revision #20110804